It is, essentially, the opposite of the mtk_get primitive. It allows
the user to change the value of a MTKObject's attribute, setting it to
a specific value.
Usage: mtk_set(tangram2_source_variable
,
mtk_object_name
,
object_attribute
[,
index_list
]
)
where tangram2_source_variable
specifies the Tangram-II variable
that holds the new value;
mtk_object_name
is the
MTKObject whose attribute will be changed;
object_attribute
the attribute's name; and [,
index_list
]
specifies the position of the attribute (if the attribute happens to be a
vector, or matrix) whose value will change.
Example: Changing the HMM's initial state probability vector (PI).
Events= event = event( TIME ) condition = ( TRUE ) action = { float new_PI[2]; ... new_PI[0] = 0.5; new_PI[1] = 0.5; ... /* Changing PI values */ mtk_set( new_PI[0], Predictor, "pi", 0 ); mtk_set( new_PI[1], Predictor, "pi", 1 ); ... };